Implementation of Artificial Intelligence Algorithms in Brain Tumor Detection and Classification

Lizhuo Xu

2024

Abstract

Brain tumors in the early stage are difficult to detect and doctors may misdiagnose due to many reasons, such as tiredness, insufficient experience, or maybe even just carelessness. Artificial intelligence (AI) algorithms can be used to help doctors diagnose. This study developed a model that identifies and classifies three types of brain tumors using Magnetic Resonance Imaging (MRI) images. This study used a dataset from Kaggle. The chosen dataset was preprocessed by first creating a validation set from the original training set. Data augmentations like brightness changing, saturation changing, and contrast changing were used. Images in the dataset were also randomly flipped. The model developed by this study used the Convolutional Neural Network (CNN) technology. Transfer learning was used in this study to promote the feature extraction ability of the model. The pre-trained model VGG19 was used as the base model. A convolutional layer and several fully connected layers were used after the base model. Dropout layer and regularizers were added to prevent potential overfitting. After training, the model showed a relatively good test performance and indicates that artificial intelligence algorithms have great potential in the task of detecting and classifying brain tumors using MRI images.

Paper Citation

in Harvard Style

Xu L. (2024). Implementation of Artificial Intelligence Algorithms in Brain Tumor Detection and Classification. In Proceedings of the 1st International Conference on Modern Logistics and Supply Chain Management - Volume 1: MLSCM; ISBN 978-989-758-738-2, SciTePress, pages 219-223. DOI: 10.5220/0013281300004558

in Bibtex Style

@conference{mlscm24,
author={Lizhuo Xu},
title={Implementation of Artificial Intelligence Algorithms in Brain Tumor Detection and Classification},
booktitle={Proceedings of the 1st International Conference on Modern Logistics and Supply Chain Management - Volume 1: MLSCM},
year={2024},
pages={219-223},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013281300004558},
isbn={978-989-758-738-2},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Modern Logistics and Supply Chain Management - Volume 1: MLSCM
TI - Implementation of Artificial Intelligence Algorithms in Brain Tumor Detection and Classification
SN - 978-989-758-738-2
AU - Xu L.
PY - 2024
SP - 219
EP - 223
DO - 10.5220/0013281300004558
PB - SciTePress